tokenization_kosmos2_fast.py

# coding=utf-8
# Copyright 2023 Microsoft Research and The HuggingFace Inc. team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
""" Tokenization classes for KOSMOS-2 model."""


import os
from shutil import copyfile
from typing import List, Optional, Tuple

from transformers.tokenization_utils import AddedToken
from transformers.tokenization_utils_fast import PreTrainedTokenizerFast
from transformers.utils import is_sentencepiece_available, logging


if is_sentencepiece_available():
    from .tokenization_kosmos2 import Kosmos2Tokenizer
else:
    Kosmos2TokenizerFast = None


logger = logging.get_logger(__name__)

VOCAB_FILES_NAMES = {"vocab_file": "sentencepiece.bpe.model", "tokenizer_file": "tokenizer.json"}

PRETRAINED_VOCAB_FILES_MAP = {
    "vocab_file": {
        "microsoft/kosmos-2-patch14-224": "https://huggingface.co/microsoft/kosmos-2-patch14-224/resolve/main/sentencepiece.bpe.model",
    }
}

PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
    "microsoft/kosmos-2-patch14-224": 2048,
}


class Kosmos2TokenizerFast(PreTrainedTokenizerFast):
    """
    Construct a "fast" KOSMOS-2 tokenizer (backed by HuggingFace's *tokenizers* library). Adapted from
    [`RobertaTokenizer`] and [`XLNetTokenizer`]. Based on
    [BPE](https://huggingface.co/docs/tokenizers/python/latest/components.html?highlight=BPE#models).

    This tokenizer inherits from [`PreTrainedTokenizerFast`] which contains most of the main methods. Users should
    refer to this superclass for more information regarding those methods.

    Args:
        vocab_file (`str`):
            Path to the vocabulary file.
        bos_token (`str`, *optional*, defaults to `"<s>"`):
            The beginning of sequence token that was used during pretraining. Can be used a sequence classifier token.

            <Tip>

            When building a sequence using special tokens, this is not the token that is used for the beginning of
            sequence. The token used is the `cls_token`.

            </Tip>

        eos_token (`str`, *optional*, defaults to `"</s>"`):
            The end of sequence token.

            <Tip>

            When building a sequence using special tokens, this is not the token that is used for the end of sequence.
            The token used is the `sep_token`.

            </Tip>

        sep_token (`str`, *optional*, defaults to `"</s>"`):
            The separator token, which is used when building a sequence from multiple sequences, e.g. two sequences for
            sequence classification or for a text and a question for question answering. It is also used as the last
            token of a sequence built with special tokens.
        cls_token (`str`, *optional*, defaults to `"<s>"`):
            The classifier token which is used when doing sequence classification (classification of the whole sequence
            instead of per-token classification). It is the first token of the sequence when built with special tokens.
        unk_token (`str`, *optional*, defaults to `"<unk>"`):
            The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this
            token instead.
        pad_token (`str`, *optional*, defaults to `"<pad>"`):
            The token used for padding, for example when batching sequences of different lengths.
        mask_token (`str`, *optional*, defaults to `"<mask>"`):
            The token used for masking values. This is the token used when training this model with masked language
            modeling. This is the token which the model will try to predict.
        additional_special_tokens (`List[str]`, *optional*, defaults to `["<s>NOTUSED", "</s>NOTUSED"]`):
            Additional special tokens used by the tokenizer.
        num_patch_index_tokens (`int`, *optional*, defaults to `1024`):
            The number of tokens used to specify the patch indices of bounding boxes in an image. These tokens have the
            format `<patch_index_xxxx>` where `xxxx` is an integer.
    """

    vocab_files_names = VOCAB_FILES_NAMES
    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
    model_input_names = ["input_ids", "attention_mask"]
    slow_tokenizer_class = Kosmos2Tokenizer

    def __init__(
        self,
        vocab_file=None,
        tokenizer_file=None,
        bos_token="<s>",
        eos_token="</s>",
        sep_token="</s>",
        cls_token="<s>",
        unk_token="<unk>",
        pad_token="<pad>",
        mask_token="<mask>",
        num_patch_index_tokens=1024,
        add_tag_and_patch_index_tokens=False,
        **kwargs,
    ):
        # Mask token behave like a normal word, i.e. include the space before it
        mask_token = AddedToken(mask_token, lstrip=True, rstrip=False) if isinstance(mask_token, str) else mask_token

        super().__init__(
            vocab_file,
            tokenizer_file=tokenizer_file,
            bos_token=bos_token,
            eos_token=eos_token,
            sep_token=sep_token,
            cls_token=cls_token,
            unk_token=unk_token,
            pad_token=pad_token,
            mask_token=mask_token,
            **kwargs,
        )

        self.vocab_file = vocab_file
        self.can_save_slow_tokenizer = False if not self.vocab_file else True

        self.eod_token = "</doc>"

        self.boi_token = "<image>"
        self.eoi_token = "</image>"

        self.eoc_token = "</chunk>"
        self.eol_token = "</line>"

        self.bop_token = "<phrase>"
        self.eop_token = "</phrase>"

        self.boo_token = "<object>"
        self.eoo_token = "</object>"

        self.dom_token = "</delimiter_of_multi_objects/>"

        self.grd_token = "<grounding>"

        self.tag_tokens = [
            self.eod_token,
            self.boi_token,
            self.eoi_token,
            self.eoc_token,
            self.eol_token,
            self.bop_token,
            self.eop_token,
            self.boo_token,
            self.eoo_token,
            self.dom_token,
            self.grd_token,
        ]

        self.num_patch_index_tokens = num_patch_index_tokens
        patch_index_tokens = [f"<patch_index_{str(x).zfill(4)}>" for x in range(self.num_patch_index_tokens)]

        if add_tag_and_patch_index_tokens:
            for idx, token in enumerate(self.tag_tokens + patch_index_tokens):
                # we need to set `special_tokens=False` to be the same as in the slow tokenizer.
                self.add_tokens(AddedToken(token, lstrip=True, rstrip=False), special_tokens=False)

    def build_inputs_with_special_tokens(
        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
    ) -> List[int]:
        """
        Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and
        adding special tokens. An XLM-RoBERTa sequence has the following format:

        - single sequence: `<s> X </s>`
        - pair of sequences: `<s> A </s></s> B </s>`

        Args:
            token_ids_0 (`List[int]`):
                List of IDs to which the special tokens will be added.
            token_ids_1 (`List[int]`, *optional*):
                Optional second list of IDs for sequence pairs.

        Returns:
            `List[int]`: List of [input IDs](../glossary#input-ids) with the appropriate special tokens.
        """

        if token_ids_1 is None:
            return [self.cls_token_id] + token_ids_0 + [self.sep_token_id]
        cls = [self.cls_token_id]
        sep = [self.sep_token_id]
        return cls + token_ids_0 + sep + sep + token_ids_1 + sep

    def create_token_type_ids_from_sequences(
        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
    ) -> List[int]:
        """
        Create a mask from the two sequences passed to be used in a sequence-pair classification task. XLM-RoBERTa does
        not make use of token type ids, therefore a list of zeros is returned.

        Args:
            token_ids_0 (`List[int]`):
                List of IDs.
            token_ids_1 (`List[int]`, *optional*):
                Optional second list of IDs for sequence pairs.

        Returns:
            `List[int]`: List of zeros.

        """

        sep = [self.sep_token_id]
        cls = [self.cls_token_id]

        if token_ids_1 is None:
            return len(cls + token_ids_0 + sep) * [0]
        return len(cls + token_ids_0 + sep + sep + token_ids_1 + sep) * [0]

    def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
        if not self.can_save_slow_tokenizer:
            raise ValueError(
                "Your fast tokenizer does not have the necessary information to save the vocabulary for a slow "
                "tokenizer."
            )

        if not os.path.isdir(save_directory):
            logger.error(f"Vocabulary path ({save_directory}) should be a directory.")
            return
        out_vocab_file = os.path.join(
            save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
        )

        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file):
            copyfile(self.vocab_file, out_vocab_file)

        return (out_vocab_file,)